The foundation that keeps the MES running while the plant keeps running

The six foundations behind VECHR MES

Configuration treated like code

Master data and production configuration are too important to move straight from development into production. VECHR MES brings promotion-pipeline discipline to manufacturing configuration.

• Dev → QA → Prod Configuration is built in Dev, validated in QA (Quality Assurance), then promoted to production once the changes have been reviewed.
• A diff before promotion A comparison between environments shows exactly what will change before the deployment happens.
• Rollback without rebuilding by hand A problematic promotion can be returned to its previous version without unpicking the changes one at a time.
• Every move leaves a trail Promotions and rollbacks are recorded per environment, so configuration changes have a history that can be traced.

Integrate over REST without writing an endpoint per call

The API Gateway is a REST facade over the same commands the VECHR MES screens use. A new capability becomes a REST route automatically.

• Zero-code routes Routes are discovered from each service at runtime, and any command beyond create, read, update and delete is reachable through a generic actions endpoint.
• API key or user token Machines authenticate with an API key and people with a bearer token. Both resolve to the same identity.
• Permissions stay where they are The gateway authorizes nothing itself. Each service applies its own permission checks, so there is no second allow list to keep in sync.
• Live OpenAPI documentation Swagger UI is generated from the current route catalog, with request bodies typed from the real validators.
• API clients with a key lifecycle Each API client has its own permission grants, and its key can be created, rotated, revoked or deleted.

API Gateway is available on the Enterprise tier.

Event-driven: one change, many reactions

Production does not wait for a dashboard to refresh. When something changes, the systems that need to know receive the event and can react in their own context.

• Changes propagate without polling A production event can update OEE, dashboards, notifications, workflows, and related services without waiting for the next batch processing run.
• Services do not have to know everything about each other Services communicate through events and clear contracts, so adding a new capability does not mean changing every service already running.
• Grows with the number of lines The same event-driven pattern applies when you add a resource, a production line, a plant, or a new site.

Reliability for a system that cannot merely be "online"

An MES is not an application someone opens when they need a report. It sits in the middle of production execution, so availability, recovery, and the ability to change things without stopping operations are part of the platform design.

Installed from one signed bundle, observable from day one

• Nothing pulled at runtime Each release is one signed file with every image and chart. It is loaded into your own container registry, and the deployment pulls nothing from VECHR or public registries afterwards.
• Works at air-gapped sites Because nothing is pulled at runtime, the same installation works at a site with no internet access.
• The same steps on every target Cloud and on-premises kits run the same phase sequence, and database migrations run as part of each deployment.
• Observability included Grafana, Prometheus, Loki and Tempo come pre-configured. A trace ID leads to the matching log lines, end to end.

Scale is not only a machine count

An MES architecture has to grow sideways as well as upward: more resources, more lines, more plants, and more transactions without changing the underlying model.

• Multi-resource New resources can be added to the plant structure without changing how the system models execution, scheduling, or monitoring.
• Multi-line and multi-area The ISA-95 structure lets new lines and areas become part of the same model rather than a separate system.
• Multi-plant Additional plants use the same platform foundation, with access control and configuration still scoped per site.
• Services can evolve independently A service- and event-based architecture lets new capabilities develop without turning the whole platform into a single monolithic deployment.

The details people notice every day

• Stored in UTC, shown in your timezone The database always stores UTC (Coordinated Universal Time). Each user sees times in their own timezone.
• English and Indonesian The interface ships in both languages.
• Light and dark Both themes follow strict contrast rules, so text stays readable on every background.
• Search and shortcuts Global search from the top bar, and a documented set of keyboard shortcuts.
• Notifications that reach the right people Notifications go out through an outbox, so they are not lost, and can target one user, several users, everyone, a role or a permission.
• Your logo on reports Generated reports carry your company branding.

AI stays inside platform governance

AI is not a special layer with a privileged route to production data. It follows the same platform rules: read the context it is permitted to read, produce an insight or a draft, then wait for a person before anything is saved.

• Read for insight AI can answer questions about KPIs, OEE, downtime, orders, batches, and lots from the MES data available to it.
• Draft for changes Routings, orchestrator flows, or master data changes can be produced as drafts for a person to review.
• Human confirmation as the gate No AI surface writes a change directly to a production record without human confirmation.
• Usage can be traced AI consumption can be tracked by user, model, turn, and period, so usage has clear visibility and ownership.